%Aigaion2 BibTeX export von HES SO Valais Publications
%Saturday 10 October 2026 10:57:21 AM

@INPROCEEDINGS{,
     author = {Mar{\'{e}}chal, Lo{\"{\i}}c},
   keywords = {asset pricing, clustering, cybersecurity., machine learning, Natural Language Processing},
      month = jun,
      title = {Disentangling the sources of cyber risk premia},
  booktitle = {Workshop on the Economics of Information Security (WEIS)},
       year = {2026},
  publisher = {Workshop on the Economics of Information Security (WEIS)},
   location = {Berkeley},
        url = {https://weis2026.econinfosec.org/wp-content/uploads/sites/13/2026/05/WEIS2026_paper_5.pdf},
   abstract = {We use a methodology based on a machine learning algorithm to quantify firms’ cyber risks based on their disclosures and a dedicated cyber corpus. The model can identify paragraphs related to determined cyber-threat types and accordingly attribute several related cyber scores to the firm. The cyber scores are unrelated to other firms’ characteristics. Stocks with high cyber scores significantly outperform other stocks. The long-short cyber risk factors exhibit positive risk premia, are robust to all factors’ benchmarks, and help price returns. Furthermore, we suggest the market does not distinguish between different types of cyber risks but instead views them as a single, aggregate cyber risk. Importantly, our cyber score captures the extent to which firms’ disclosures are semantically related to known cyber attack descriptions and therefore reflects exposure, awareness, or discussion of cyber risk, rather than realized cyber incidents. We provide evidence that the score is predictive of future cyber-related disclosures, supporting its interpretation as a proxy for latent cyber risk.}
}